yt-packaging
Pass
Audited by Gen Agent Trust Hub on Apr 4, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection surface detected. The skill ingests untrusted data from a local file (video-brief-yt-packaging.md) and external YouTube search results via the WebSearch tool. It lacks explicit boundary markers or delimiters to isolate this external data from the agent's core instructions. The capability inventory includes the Write tool, which allows the agent to modify files in the working directory (e.g., saving packaging-yt-packaging.md). No sanitization or validation of the input data is described. This structure creates a risk where malicious instructions embedded in a video brief or a search result could influence the agent to perform unauthorized file modifications or deviate from its intended task.
Audit Metadata